Why 3D-printed parts are weaker than they look
A printed part is a stack of welded layers, not solid plastic — and the weld between layers is its weak point. Why orientation is the most important decision in 3D printing, and how much strength you actually lose.
A printed part is a little laminate
An FDM printer builds a part by squeezing out molten plastic in thin roads, side by side to fill a layer, then stacking layer on layer. Within a single layer the plastic is continuous and strong. Between layers it is not: the bond forms when a fresh hot road partly re-melts the cooler one beneath it — and that weld never fully heals. Small voids and weaker joins live at every layer interface.
So a printed part behaves like a laminate, not like solid moulded plastic. It is strong in the plane of the layers and weak across them, in the build (Z) direction. That single fact explains almost every surprising 3D-printing failure.
How much you actually lose
Measured interlayer strength in the Z direction is typically only 30 to 70 percent of the in-plane strength — and it can be worse with poor settings: a cold nozzle, fast printing, or a draught cooling each layer before the next bonds to it.
The exact number depends on the material, the temperature, the layer height and the speed. That dependence is the real lesson. A printed part does not have a single 'strength'; it has a strength that depends on how it was made and which way you load it. This is the printing version of the biggest idea in engineering — the gap between a datasheet value and what a real part holds.
Orientation is the design
Because the weak direction is always across the layers, how you orient the part on the plate is the single biggest strength decision you make — before infill, before material, before anything.
Lay the part down so that tension and bending run along the layers, never across them. A hook printed lying flat, with the layers stacked across the pull, can snap at a small fraction of the load of the same hook printed standing on edge, with the layers running along the pull. Same printer, same plastic, same settings — only the orientation changed.
What helps, and what the other processes do
A hotter nozzle and an enclosed, warm chamber give the layers more time and heat to weld, which is exactly why high-performance printing (PEEK, PC) needs a heated chamber. And it is the walls — the solid perimeters — that carry the load, not the infill; beyond about 40 percent infill you are mostly adding time and weight.
The other additive processes are less anisotropic. Resin (SLA/DLP) parts cure by cross-linking that reaches across layers, so they are nearly isotropic — but the resin is often brittle. Powder (SLS/MJF) parts are the most isotropic of all, which is part of why they are chosen for functional end-use parts.
Design to the part, not the datasheet
A material page will tell you the tensile strength of bulk PLA. A PLA part pulled across its layers holds a fraction of that. Neither number is wrong — they measure different things — but using the first to design the second is how printed parts fail in the field.
The honest move is to know the direction of your loads, orient the part so the layers carry them, and treat the through-thickness strength as the weak link it is. That mindset — the real number, not the ideal one — is the whole point.
1. Why is a 3D-printed part weakest across its layers?
2. A hook must carry a downward pull. How should it be printed for strength?
3. Typical Z-direction (interlayer) strength is roughly what fraction of the in-plane strength?